The value of magnetic resonance imaging in the evaluation of fatty filum terminale. Academic Article uri icon

Overview

abstract

  • OBJECTIVE: To determine whether there are magnetic resonance imaging (MRI) characteristics of fatty fila that are correlated with neurological deficits, especially in the presence of a normal-level conus medullaris. METHODS: Lumbosacral MRI scans were reviewed for patients with fatty fila who were treated at Duke University Medical Center during a 5-year period. The patients were divided into three groups. Group I patients (n = 5) had fatty fila that were incidentally detected during evaluations for metastases or infections. Group II patients (n = 16) exhibited isolated low back pain but were in neurologically intact condition. Group III patients (n = 15) exhibited neurological impairments consistent with distal spinal cord dysfunction. Several characteristics were measured on the MRI scans, including the location of the conus medullaris, the filum thickness, and the distance of fat from the conus. These results were assessed for statistically significant correlation with the presence of clinical symptoms. RESULTS: The majority of patients in all three groups demonstrated the normal conus position (L2 or above) and thickened fila. The distance of fat from the conus was the only parameter that demonstrated a statistically significant difference among the groups. CONCLUSION: The following findings were noted: 1) patients were likely to exhibit neurological deficits at a younger age (<22 yr in Group III versus 47 yr in Groups I and II); 2) a conus level below L2 was associated with neurological deficits (Group III); 3) filum thickness was not correlated with clinical presentation; 4) fat in the filum within 13 mm of the conus medullaris was most predictive of neurological deficits (Group III).

publication date

  • February 1, 2004

Research

keywords

  • Adipose Tissue
  • Cauda Equina
  • Magnetic Resonance Imaging
  • Nervous System Diseases
  • Spinal Cord

Identity

Scopus Document Identifier

  • 10744223392

Digital Object Identifier (DOI)

  • 10.1227/01.neu.0000103451.63301.0b

PubMed ID

  • 14744284

Additional Document Info

volume

  • 54

issue

  • 2